You are on page 1of 6

Applying Intelligent Transport Systems to

manage noise impacts

Isabel Wilmink
TNO & TrafficQuest, Delft, The Netherlands.
Tanja Vonk
TNO, Delft, The Netherlands.

This contribution discusses how traffic management, and many other measures that can be
categorised as Intelligent Transport Systems (ITS, i.e. all traffic and transport measures that use
ICT) can help reduce noise levels by influencing mobility choices and driving behaviour. Several
examples of such measures and how they impact noise levels will be given, ranging from giving
travel, route or departure time advice to automated driving (“self-driving vehicles”).
PACS no. xx.xx.Nn, xx.xx.Nn

1. Introduction 2. Choices of travellers

Cities are growing and this means that urban road From the research reported in [1], it follows that to
and public transport networks are going to be more reduce noise levels along a specific road, traffic
heavily used as well. Motorway networks also and transport measures should be mainly aimed at
show increased usage. Although the accessibility achieving changes in:
of many regions and cities is still improving, there 1. vehicle speeds – optimizing both the
are also drawbacks to increased mobility: average speed and the highest speeds
liveability and environmental problems (and traffic (especially at night time);
safety, as well, but that is out of scope for this 2. traffic volumes and composition –
paper). specifically so that the noisiest vehicles are
In the field of traffic and transport, attention has banned or given an incentive to take an
shifted from measures aimed at motorway traffic alternative route;
to measures aimed at urban areas. This also means 3. acceleration/deceleration – avoiding
that there is a shift in what those measures aim to extreme accelerations (and sometimes
achieve: from a focus on travel times (and perhaps even decelerations, e.g. on slopes)
traffic safety, depending on local traffic safety
statistics) to finding a balance between those Road users tend not to think in these terms though
objectives and objectives regarding liveability and – they make travel and driving behaviour choices
pollution. The question addressed in this paper is: as described below, and those choices can be the
how can road traffic and transport measures help starting point for the design of measures to reduce
reduce noise levels / noise annoyance? This noise production.
contribution discusses how innovative traffic Figure 1 gives an overview of all the choices that
management, and many other measures that can be travellers can make (and the figure also shows that
categorised as Intelligent Transport Systems (ITS, there are interactions between these choices – a
i.e. all traffic and transport measures that use ICT) choice on one level influences choices on other
can help reduce noise levels by influencing levels). The choices range from long-term choices
mobility choices and driving behaviour. Several such as where to work and where to live to very
examples of such measures and how they impact short term driving behaviour choices such as how
noise levels will be given, ranging from giving aggressively one accelerates at an intersection.
travel, route or departure time advice to automated Most of the choices are made pre-trip, i.e. at home
driving (“self-driving vehicles”). or at the office, but some choices are made on-trip
and can, with many drivers having navigation

Copyright© (2015) by EAA-NAG-ABAV, ISSN 2226-5147 1053

All rights reserved
I. Wilmink: Applying... EuroNoise 2015
31 May - 3 June, Maastricht

Figure 1. The traveller’s pre-trip and on-trip choices.

devices or smartphones in their vehicles, be such that they are during the most
influenced real-time. favourable time periods, noise-wise. Here,
Examples of how these choice can work out are: there might be a clash with the aim of
• Location choice: Land use choices (and many traffic measures currently in place to
regulation) can help reduce commuting influence departure time, as they are
distance, can influence what routes are generally aimed at having car (and truck!)
used by what types of vehicles (e.g. no drivers avoid driving during the peak
noisy vehicles in residential areas), and periods, and that often leads to a shift of
could influence whether or not one can departures to earlier in the morning, which
drive a silent electric vehicle (including means more traffic in the “night time”
ones with limited ranges). (before 7 in the morning). There is also a
• Trip (frequency) choice: To make the trip link with encouraging employees to work
or not. If employers allow or even from home – in this case, perhaps, only for
encourage employees to work from home, a few hours in the morning.
this could lead to fewer early morning • Route choice: Some routes cause more
commutes. problems, in terms of noise annoyance,
• Destination choice: where to travel. And than others. With more and more travellers
that can be anywhere now, with many using traffic information and navigation
destinations presenting themselves on the devices, and governments opening up their
internet and navigation systems making it data about traffic conditions (real-time),
easy to find them. there are also more and more possibilities
• Mode choice: Obviously some modes are to influence route choice. From a noise
noisier than others. Travellers can be perspective, it could be worthwhile to get
encouraged to use less noisy modes, if the in touch with service providers to ensure
benefits for them are clear. (This does not that noise levels are a criterion in their
include the choice to purchase a silent route advices (‘social navigation’).
vehicle, e.g. an electric scooter, bicycle or • Driving behaviour: This is an area that is
car, but incentives to do that can certainly starting to get interesting, as functionalities
help, because once a silent vehicle is are becoming available that can
available it can be used.) significantly influence driving behaviour,
• Departure time choice: From a traffic including choices regarding speed and lane
demand management point of view, one use, following behaviour,
would like to influence departure times acceleration/deceleration behaviour and

EuroNoise 2015 I. Wilmink: Applying...
31 May - 3 June, Maastricht

gear choice. Vehicle automation is a hot • Improving bicycle facilities

topic and the driving strategies of

automated vehicles (“driverless cars”) 3.1. Real-time route advice
could be designed to minimise noise There have been many advancements in data
production (taking into account collection and processing over the past years,
information about the areas along roads). which means that there is more information about
Automated vehicles could drive even how and where people travel than ever. Also, a
smarter by using information large share of travellers is connected to the
communicated to them from other vehicles internet, through their smartphone, navigation
or road-side and back office applications. device or in-car systems. That means that there are
This could be used to achieve a better many ways to for travellers to get advice on what
distribution of traffic over the network, or is the best way to travel for them.
even over the lanes of a specific road In traffic management, a distinction is made
section. between pre-trip and on-trip information. Pre-trip,
travellers research their destination and the way to
The next paragraph discusses some innovative get there (mode and route choice, departure time,
measures that could be used to combat noise parking or P+R advice). Websites and apps
annoyance, in their present form or in an adapted combine historic and real-time data to estimate the
form, with knowledge about how the measure travel time via various routes so that they can
could help reduce noise levels incorporated. advise on the best route at that time. The advice
can be updated on-trip as well these days, which
3. Innovative traffic measures makes travellers more flexible.
It is not likely that noise annoyance is a factor in
Traffic and transport measures used to be mostly the advices generated by websites, navigation
physical measures, but these days there is not systems and smartphone apps. Mostly, the advice
much room, or budget, to add or enhance physical given is the fastest route at the time that the
infrastructure. Thus, the emphasis is on better use travellers has indicated he/she wants to travel.
of the existing infrastructure through many There are some systems that offer the possibility to
different ITS measures – fitting with the idea of choose the most fuel efficient route. The ‘most
Smart Cities. Information about what are the most liveable’ route could be added as well – the
promising ITS measures is becoming available, information needed to generate such route is
using results from field tests and widespread available, though probably not used for this
implementations of some more mature measures. purpose yet. And travellers may be willing to
If we can inform travellers of the alternatives they choose a more liveable route if the difference in
have for their trips (pre-trip and on-trip), they can travel time is small. Such ‘social navigation’ has
make better choices (or their in-veihcle systems been researched (see e.g. [2]) and can be quite
can do that for them). Impacts on noise will need effective, depending on the level of altruism that
to be compared to impacts related to other policy travellers display. Adding an incentive like a
goals cities have (e.g. accessibility/travel times, reward for choosing a liveable route could be
traffic safety, air quality). Decision support models considered – see section 3.2 for a discussion of
that enable multi-criteria optimisation can be how financial incentives have been shown to
deployed to make it easier for road authorities, substantially influence drivers’ choices. In
traffic operators and service providers (e.g. of addition, there are large scale field trials with apps
travel advice apps) to decide how and when a that give route advice and much can be learnt from
measure should be applied. these trials about the user acceptance and
In this section, the following innovative traffic compliance with advices from these trials. For
measures are discussed: instance, in the Amsterdam Practical Trial [3],
• Real-time route advice apps are currently being tested that give different
• Incentives to avoid driving in peak periods route advices to different types of travellers. The
• Network-wide traffic management app takes into account the state of the network,
• Traffic control with the aim to distribute traffic more efficiently
• Automation of the driving task over the network (in this case, seeking to reduce

I. Wilmink: Applying... EuroNoise 2015
31 May - 3 June, Maastricht

delays). The travellers’ trips are logged so their accept the measure (but a reward certainly helps).
compliance with the route advices can be analysed, Also, along with the request to not use particular
and travellers are asked to fill in questionnaires, roads, advice could be given about alternatives:
which gives insight into the motivations of working from home for a few hours before going
thousands of travellers in the Amsterdam area for to the office, or using quiet and clean modes of
using such apps. transport such as electric scooters and bicycles.

3.2. Incentives to avoid driving in peak periods 3.3. Network-wide traffic management
There are several ways to influence the departure Network-wide traffic management is based on the
time choices of travellers. In several countries same objectives as real-time route advice, but
pricing measures have been implemented, like tolls looks at it from the network perspective (and thus
or a congestion charge. The technology to collect often from the road authority’s perspective) rather
tolls has, in many countries, advanced to a point than the perspective of the individual traveller.
where it is possible to charge variably, with higher Now that attention has shifted to urban traffic
rates during peak hours. In theory, the rates management, attention has also shifted from
charged could also be made to depend on the accessibility to liveability, or rather to finding a
impact of trips on the liveability in an area. In balance between multiple objectives.
practice, there might be an adverse effect of Important instruments in urban traffic management
pricing measures that are implemented to are controlled intersections (traffic lights), variable
discourage travelling in peak periods – more trips
in the very early morning, for instance, in very
congested networks.
In the Netherlands, road pricing or congestion
charges aren’t popular measures. Instead, several
measures have been implemented that reward
travellers instead of charging them extra for the
use of infrastructure. They are labelled as ‘Peak
traffic avoidance’ measures (‘Spitsmijden’ in
Dutch). An overview report of several Spitsmijden
projects showed participants in the projects
(ranging from a few hundred to over 10,000 per
project) changed their behaviour such that they
avoided 30-60% of their peak period trips [4] –
with some participants continuing this behaviour
even after the reward stopped (with the reward
being a few euros per peak period avoided). Figure 2. Noise map as used in the The Hague DSS.
The original intent of measures rewarding
travellers who avoid peak period traffic is to better
spread traffic volumes over time, to prevent message signs and parking guidance systems. All
breakdown of traffic. Such measures are especially can be optimised for noise reduction. For parking
useful during road works or other temporary guidance, it is even relatively easy to favour silent
situations, where large numbers of travellers need vehicles by reserving spaces for them or offering
to be persuaded to change their travel behaviour. reduced rates.
License plate cameras or in-car devices are used to Influencing traffic flows using traffic light control
identify possible participants and determine and variable message signs is more complex but
whether vehicles have indeed stayed away from has been tried out in The Hague. For a part of the
certain roads during peak periods. Such measures road network of this city, a multi-criteria decision
could also be considered for roads where noise support system was built. In the Decision Support
levels need to be reduced – mostly as a temporary Tool (DSS) the situation in the network in terms of
measure (until more permanent changes can be noise, pollutant emissions and traffic performance
made). The measures would need to be explained is presented with graphs, maps and a grade [5][6].
very well to road users in order to get them to Also, the expected impact of measures can be

EuroNoise 2015 I. Wilmink: Applying...
31 May - 3 June, Maastricht

calculated. This is used to inform drivers of This is not the focus of automotive engineers right
preferred routes and is supported by adaptation of now, but stakeholders with expertise in the area of
the traffic signals, so that it becomes more noise mapping and noise abatement measures
attractive to use the preferred routes. Instead of could get involved. There is link with other vehicle
just using travel times and queue lengths as measures reducing noise, like quieter tires – when
triggers, now also air quality and noise levels can innovating one aspect of vehicles, other aspects
be triggers. could be innovated as well.

3.4. Traffic control 3.6. Improving bicycle facilities

The previous section discussed network-wide Cycling is currently promoted in many cities, as it
traffic management, but management of single is a very sustainable and healthy mode of
intersections or a string of intersections can also be transport. The sales of electric bicycles are
beneficial. booming and this means that more and more
Traffic control is becoming more flexible and travellers can ride longer distances than they
better informed – in the sense that the algorithms would on a regular bike – making it possible to
used receive more and/or better information about commute by bike, for instance. But even riding a
what traffic is arriving at an intersection, which regular bike can be encouraged by providing good
means that they can optimise which directions get facilities (such as supervised bicycle parking).
green and for how long. In the future, Internet communities and bicycle navigation apps
communication between vehicles and traffic lights can be used to encourage travellers to make more
will make the information better still – the sustainable mode choices. A good way to
vehicle’s navigation systems can also inform familiarise people with bicycles is by introducing
traffic lights about the direction that the vehicle smart bicycle renting systems in cities (e.g. Velib'
will take (turn left or right, or go straight ahead). in Paris, Villo! in brussels). Even bike-sharing has
And in turn, the traffic lights can inform vehicles become possible with apps bringing bike owner
about the best lane and speed to choose. This will and client together.
result in fewer stops and less acceleration
behaviour. If information about the vehicle type is
included, priority can be given to noisy vehicles 4. Monitoring and evaluation
(for a short term effect: they produces less noise if
they keep driving compared to stop-and-go Traffic monitoring is not a measure as such, but
behaviour. For the longer term, giving priority to helps to provide input to measures, so that suitable
the noisiest vehicles is less desirable, as this does strategies and advices can be generated. Also,
not encourage a shift to less noisy vehicles). monitoring is needed to generate data for
evaluation of the effectiveness of measures.
Traffic monitoring anno 2015 uses various sources
3.5. Automation of the driving task of information, and combines them using data
Automation of the driving task (resulting in fusion techniques [7]. The resulting information is
‘driverless cars’) is a hot topic and many benefits increasingly shared as open data – between road
are expected from it. It will take a while before authorities but also with service providers who can
drivers are really not needed anymore, but vehicles use the information to generate traffic information
are already taking over some driving tasks even in and travel advice. Sensors along or embedded in
the average car of today (cruise control) and will the road provide large amounts of mostly quite
start taking over more and more tasks. Vehicles are accurate data about traffic volumes and speeds,
going to be programmed to drive safely and and sometimes traffic composition. Also, probe
comfortably (right now, much like how humans data from vehicles finds its way to more and more
drive). They could also be programmed to drive applications. Probe data can be collected
with as little noise as possible – much of the everywhere in the network, and can give quite
information needed for algorithms optimizing accurate (in space and time) data about speeds.
performance in terms of noise production is Information about traffic volumes and/or traffic
already available in digital maps, and more composition is less accurate, given the still low
information that could be relevant can be added. penetration rates (order of magnitude: 2-4% of

I. Wilmink: Applying... EuroNoise 2015
31 May - 3 June, Maastricht

vehicles on the road). However, the number of cooperation with Connekt/ITS Netherlands, available @
vehicles serving as probes will increase, and with netherlands-progress-report-2014-def.pdf.
that the data will become more useful over the [4] MuConsult: Eindevaluatie Mobiliteitsprojecten.
coming years. Dedicated apps may be developed Amersfoort, MuConsult, 2 November 2013 (in Dutch).
and used to also measure other useful data such as [5] Vonk, T., , P. vd Mark, Y. Chen, T. Basten, P. Wessels,
accelerations and other relevant events. E. Kuiper, Z. Woldeab & A. Eijk: Duurzaam
Before deploying a measure, traffic simulation Dynamische Verkeersmanagement in de Gemeente Den
Haag. TNO, October 2013 (in Dutch)
models can be used, off-line or real-time, to
evaluate what order of magnitude of traffic effects [6] Vonk, T. & A. Reijneveld: TEN effecten in Den Haag,
Duurzaam Dynamisch Verkeersmanagement door
measures will have. Depending on the type of integrale afweging van Traffic, Emissie en Noise (TEN)
measure, and what travel choices are influenced, a bij inzet van scenario’s, Contribution to the Nationaal
microscopic (modelling individual vehicles and Verkeerskunde Congres, November 2012 (in Dutch).
drivers) or macroscopic (model traffic flows at the [7] Wilmink, I. D. Vonk Noordegraaf, T. Thomas, V.
Westerwoudt & M. Barten: Datafusie Pilot – Openbare
link level) is used, see also [1]). resultaten, NDW & DITCM, 20 October 2014 (In
Dutch; English summary available by contacting NDW
5. Conclusions

Intelligent Transport Systems (ITS) can be used to

reduce road noise. At the moment, the focus of ITS
measures is on improving accessibility and safety,
and energy efficiency (fuel costs) is also a trigger
for some measures. Noise does not seem to be on
the agenda as much. But in designing ITS
measures, noise can certainly play a role and
expected noise impacts can be weighed against
other impacts.
It has become easier to reach travellers (through
smartphone apps and the internet, but also through
roadside message signs). Influencing their travel
behaviour has become easier too – if the travellers
understand the reason or can be shown clearly how
they benefit from it. Some measures implemented
over the last decade have shown that behaviour
can indeed be changed substantially. Including
noise as a trigger for measures can mean that
measures have to be designed differently. This
requires a dialogue between different stakeholders,
e.g. road authorities, environmental departments
and service providers.

[1] IMAGINE: Guidelines for the use of traffic models for
noise mapping and noise action planning. IMAGINE
report IMA02DR7-060531-TNO.10, 31 May 2006.
[2] Bosch, A. van den, B. van Arem, M. Mahmod & J.
Misener: Reducing time delays on congested road
networks using social navigation. Proceedings of the
2011 IEEE Forum on Integrated and Sustainable
Transportation Systems, Vienna, Austria, June 29 - July
1, 2011.
[3] Connekt: ITS in the Netherlands – Progress report 2010-
2014, Ministry of Infrastructure and the Environment in